Diagnosing Ecohydrologic and Atmospheric Feedbacks from Beetle-Induced Tree Mortality with a Coupled Atmosphere-Hydrology Model

Headwater ecosystems in the Rocky Mountains of Colorado, which are paramount to domestic and agricultural hydrologic resources, are dramatically changing due to infestations by the mountain pine beetle (MPB), Dendroctonus ponderosae. Documented and modeled hydrologic responses to outbreaks include cessation of overstory transpiration, local increases in soil moisture, changes in snow accumulation and ablation, and differences in groundwater and runoff contributions to streamflow. Watershed response to infestation is, however, often inconsistent and damped at large scales, a fact often attributed to scale-sensitivity, spatial variability, and mitigating processes, especially simultaneous forest recovery. Potential atmospheric feedbacks have not yet been thoroughly investigated as a possible compensating factor, despite modeled evidence that changes to latent and sensible heat flux at the surface could propagate into the atmosphere in the form of increased surface temperatures and planetary boundary layer height. Here we present controlled numerical experiments that resolve complicated feedbacks from land disturbance to atmosphere, using the Weather Research and Forecasting (WRF) mesoscale meteorological model. WRF is coupled to ParFlow, a physically-based, integrated hydrologic model, through the land surface model Noah. The model was run at high meteorological resolution, 1-km lateral grid spacing, over the Colorado headwaters region. Vegetation parameters for evergreen needleleaf trees were adjusted to reflect beetle-induced reductions in stomatal conductivity and LAI, and an ensemble methodology was used to represent a measure of uncertainty in initial atmospheric conditions. Results suggest that MPB signal does result in perturbations to atmospheric moisture, stability, and even precipitation. However, atmospheric responses are inconsistent and often insignificant when compared to ensemble spread. Changes to the land surface energy budget and to ground and near-surface air temperatures are damped when compared to meteorological models that lack a lateral flow hydrology component. This work presents the applicability of a deterministic, integrated climate-hydrologic model to identify complicated physical interactions occurring with forest disturbance, which may not be discernable with simpler models or observations.